A block-wise MAP decoder using a probability ratio for branch metrics

In this paper, we propose several methods for the block-wise MAP decoder for turbo codes. The novelties of this method are a branch metric obtained from the logarithm of the probability ratio and a pipelined and time-shared structure. The proposed block-wise method can reduce the excessive memory requirement and the decoding delay without increasing the complexity. In the simulation, the effects of quantization, the limits of input signals and the approximation of Max operation are considered.